General.Zhukov

On 5/25/2026 at 1:42 AM, opsyscons said:

I would love an example of these flexibilities that didn’t exist before and actually improve the quality of our acquisitions? I need the morale boost.

Less Flexibility:

• Probably the biggest new RFO tool being de facto prohibited, and the post-RFO environment being more restrictive than the old FAR---well, that was quite the rug pull.

• In past 12 months, mandates have added on around five brand-new >HCA approvals.

• The most draconian mandate is yet to be issued (OFPP-designated required use contracts). A very broadly scoped set of required-use contracts would be like the introduction of smallpox into the New World of agency contracts, or effectively resurrecting EO 14240 (All to GSA), or [chose your own comically overdramatic metaphor], so the parties involved have wisely not rushed this one.

More:

• The second tier BPAs are a great tool for a niche use case. (They sure look plainly illegal to me - all contractors awarded such contracts shall be provided a fair opportunity to be considered- but I am not a lawyer) Super useful when you have a multiple-award IDIQ as a potential source, but competing every new requirement is too much competition, and a single order (with lots of options) is too little competition.

• Off/on ramps are great.

• The clause clean up isn't exactly a flexibility, but it very welcome and is probably much more useful to contractors than to workforce.

• Since <5% of contract actions are FAR 15 outside of DoD, I am skipping that part since its mostly irrelevant (I'm joking, sort of).

This is what comes to mind. Probably more, right?

On 5/8/2026 at 12:07 PM, Retreadfed said:

My question is who s behind this and why?

RFO :This update [to FAR Part 16] represents a deliberate shift from a restrictive to a permissive framework, empowering contracting officers to use novel and innovative contract structures ..."

Seven months later...

EO: Use of any non-fixed-price contract...must be justified in writing by the contracting officer to the agency head.

Who - well, probably not the RFO team. and also probably not someone familiar with how governments actually work, unlike these two:

Jennifer Pahlka, whose work I admire: The response to every failure is a new layer of oversight and approval.
James Q Wilson, whom everyone should admire: The response to any scandal or failure is to add another layer of oversight.

23 hours ago, FrankJon said:

I see little cost risk where the agency dictates the number of hours it wants and then allows the contractor to work up to that ceiling. There is, however, performance risk

Agree with FrankJon generally here from my view down in the trenches.

Anecdote:

I just had a talk this week with an office the has a LH technical support desk contract that will be soon converted to FP (this conversion pre-dates the EO). Their two initial concerns were that they couldn't estimate accurately enough the workload of the help desk to convert to FP, and that FP just means more expensive in exchange for nothing. I think they would argue that the performance risk is the government's and can't be transferred- customers don't know and don't care about the employer of the help desk rep. They see their cost risk as lower under LH, since their expressed cost risk was having to spend more money on their help desk. FP means either price is too high - they spend more for the same thing - or the price is too low, contractor will cut corners to save money and that will reflect poorly on their office, not the contractor and that will also lead to higher prices later on. So either way they will lose.

For what it's worth, I think they are wrong on all accounts, but that's what I heard.

Also - "Government in Fiscal Year 2024 identified approximately $120 billion obligated on cost-reimbursement consulting contracts alone. " I looked this up with FPDS, and I don't see how this number is possible. Non-FP contracts for all services - not just cost and not just consulting - is $189 billion. I see no way to slice the data to get to their result from public data using standard definitions.

EO

(a) To the maximum extent consistent with law, and except as provided in subsection (b) of this section, executive branch departments and agencies (agencies) shall, in procurement, utilize fixed-price contracts, which for purposes of this order shall mean fixed-price contracts as defined in Part 16 of the Federal Acquisition Regulation, codified at title 48, Code of Federal Regulations, or contracts that tie profit to performance-based metrics when appropriate.

(b)(i)  Use of any non-fixed-price contract, including a cost-reimbursement contract, a time-and-material contract, a labor-hour contract, or any other non-fixed-price type of contract under Part 16 of the Federal Acquisition Regulation, must be justified in writing by the contracting officer to the agency head. 

(ii)   If the value of a non-fixed-price contract...exceeds [$XMM], then the agency head must approve the contract in writing:

(iii)  Agency heads may delegate approval under subsection (b)(ii) of this section to appropriate non‑career employees within the agency.

(iv)  Subsection (b)(ii) of this section shall not apply to contracts that [support disaster, contingency, R&D, major systems, FAR 34, FAR 35.]

What is the steel man argument here? What is the strongest case for why this EO is a good idea?

Am I not understanding something - does this imply "justified in writing" but not approved under threshold? If so, what does this mean?

FPDS has been replaced with:

usaspending.gov, which shows the FPDS data with a modern interface. It has keyword search. You can export results as a file.

sam.gov has a much more powerful FPDS report builder. I have a GVT account, not familiar with the public version.

Here is me searching for the keyword 'spaghetti' on usaspending.gov. This also works with partial contract numbers.

1 hour ago, formerfed said:

An interesting study would involve seeing how many orders versus formal contracts are done by 1102 contract specialists.

I can answer that question. For civilian agencies, Orders are the vast majority, but as you suggest, this distinction is no longer as important as it once was. What once were C contracts are now orders. We have so many GVT-wide contract vehicles - GSA lists 381 of them. Thousands of internal-use-only IDIQs and BPAs. The need to make 'C' contracts is increasingly rare - FAR 15 is the last resort.

Consider 75FCMC23F0133 Current Sources of Income and Employment Verification Services. It provides a critical service that is public-facing, complex and high-stakes. Many millions of Americans applying for federal benefits every year rely on it. If it broke it would make the news. It obligates a bit under $100MM per year. Ceiling is $2B over 5 years, First RFI went out more than year before the RFQ was issued. The largest action its agency awarded that year. Evaluation was long and complicated. It is a FAR 16.505 order. It's not on sam.gov.

6 hours ago, Vern Edwards said:

The CO job would be vey different than it is today, more attractive, and higher-paying.

The SEC has about 40 people in the contracting occupational series whose base pay is more than $200,000. All 20 contracting personnel at the Federal Reserve makes more than $180,000. At least 18 at the FDIC make over $250,000. How are things going in these agencies who clearly aren't bound by normal GS pay scale constraints? These are tiny offices, I grant you, but they would be an interesting study in what happens when 1102 salaries are much higher.

My training is in economics and I'm somewhat familiar with the (less mathematical) literature. Economic papers on this topic emphasize that there are differing markets conditions:

• low-complexity vs. high-complexity requirements

• importance of price vs. quality

• Homogenous or heterogenous sellers. (when everyone proposes the same solution to your problem; there is little to be gained by negotiating)

• Irregular/unpredictable or frequent/predictable purchasing

• Many others.

Buyer likewise have a set of procurement methods - the most important of which is using either negotiation or competitive bidding. Note that in this literature buyers using both competitive bidding and negotiation - receiving many competitive offers and then conducting negotiations - is usually waved away as a self-evidently poor choice. Clearly economists have not read and absorbed FAR 15. I'm only half-joking here - I mean FAR 15 as its understood and used by practitioners, not by FAR originalists.

Given all this, the answer to the original question 'Is it true?' is 'It depends.' When the buyer (GVT) uses the right methods for that market's conditions, then Yes costs decrease and quality increases. When the wrong method is used, then the inverse is true.

The government treating all contracts as transactions suggests the government is using competitive bidding when the conditions favor negotiation - that's when the answer is no.

P.S. A department is doing a natural experiment on this right now - many requirements whose conditions suggest (to me) they are negotiations, and previously were (more or less) negotiated, are now being treating like transactions. Maybe I'm wrong and it'll work out.

P.P.S. 'Competition under Incomplete Contracts and the Design of Procurement Policies', Carril et al, 2024, is a recent article about this topic.

On 2/20/2026 at 9:21 AM, DaveShaw said:

Is there a regulation that does not limit period of performance to 12 months under economy act order?

The servicing agency is probably wrong here if they are telling you that your Economy Act order PoP cannot exceed 12 months, even though the 1) funds are multiple-year appropriations and 2) the services are non-severable. Presumably, this both a misapplication and misinterpretation of the bona fide need rule. Usual caveats apply: A definitive answer depends on the exact situation. This is a guess.

Generally: For a typical Economy Act order, funding limitations (such as the bona fide need rule) generally are based upon the funds used, and funding rules and rules determination are the responsibility of the requesting activity, not that of the servicing activity. Although the servicing activity does have to ensure the requesting activity isn't attempting to circumvent the law.

Also note: My agency routinely does Interagency Agreements under the Economy Act using non-appropriated or no-year funds (so not identical to you) and these IAAs are exempt from the bona fide need rule - they can fund severable services for > 12 months, or fund in current FY severable services which will start in the next FY. These funds have many rules and limitations, but that isn't one of them.

12 hours ago, Don Mansfield said:

Keep in mind FAR part 13 permits use of any appropriate combination of the procedures in parts 13, 14, 15, 35, or 36.

As a practitioner, I view federal acquisition as having risk tiers. The Simplified Acquisition Threshold (SAT) serves as the primary—though not sole—demarcation for low-risk acquisitions. Removing this limitation would allow low-risk procedures to be applied to acquisitions that would traditionally be categorized as medium or higher risk due to their dollar value. Scenarios:

1. High $ and Low Risk. Much is gained (speed, simplicity) without much or any downside. Nothing is wrong with this.

2. High $ and Medium+ Risk. If the risks are known and dealt with appropriately - which means doing more scrutiny and risk mitigation that is required by SAP, aka using an appropriate combination of procedures in 13, 14, 15, 35 or 36 - nothing is wrong, I think.

3. High $ and Medium+ Risk. If the risks aren't known and/or aren't dealt with appropriately - bad.

What could go wrong depends on the decisions of COs deciding what to do with an acquisition that is high $ and probably needs more scrutiny than doing only what is required under SAP. Are they going to listen to the little angel on their right shoulder and go with #2, or listen to the devil and go with #3? What are their incentives?

Also, there are other and better ways than $ and FAR Parts for appropriately identifying and mitigating risk.

Finally, I suspect the impact - or what could go wrong - about unlimited SAPs is very different for DoD vs. everyone else. DoD does perhaps twice as many actions >$9 million than all other agencies combined. So for the DoD upside of removing the limitation is probably more than for anyone else (they would take advantage of it far more), and the potential downside might be less since DoD (presumably) has scale-dependent controls and risk-mitigation procedures that civilian agencies just don't have.

Sure, close it. I wanted to read what you all think, and I have.

Good ideas here.

Given a critical shortage, I wonder about how I would distribute the not-enough COs were I the SPE. (Note if the actual HHS SPE reads this: Hi!) Collect them all up to pool resources? Would that help? That is centralization, and what sort of stuff is best suited for this transactional arms-length method in these circumstances? (GSA seems to think the answer is everything. The end users I talk to every day disagree.) The many esoteric and unique things federal agencies do - you cannot centralize that. But the rest of acquisition - the mundane stuff that is the majority of the work - not sure. The classic which is more important: Responsiveness to end-user vs. economies of scale. How does this balance work when you've lost half your workforce? Are there economies of scale in federal acquisition? It's not obvious to me if COs are more like barbers (no economies of scale) or bankers (huge economies of scale). - is there some benefit to gathering up not-enough COs and putting them in one place? With a critical shortage, it's not obvious to me what's the best response.

I'd find the many unique things that HHS acquisitions do and make sure that at least 3 people - including at least one under the age of 50 - know how to do it. Inevitably, there will be unique things that nobody remaining knows how to do, and I'd want to find those gaps and plug them.

Also, day one I'd shanghai anyone I could into COR duty like an old-time Royal Marine roaming the wharfs of Liverpool.

HHS did, in fact, lose about 40% of its contracting workforce last year. The contracting requirements have not changed. Highlights:

• HHS lost the most 1102s of any department or independent agency.

• GS-7/9/11: 65% loss across HHS.

• The big three spending agencies within HHS (each obligates ~$8 billion per year), Center for Medicare and Medicaid, National Institute of Health, and Center for Disease Control: >50% loss

• For contrast: DoD: -9% (2,500 FTEs) and all civilian agencies: -20% (3,200)

You are in charge, what do you do?

P.S. You have also lost half of your certified CORs.

GSA re-used a previous solicitation for a very similar travel agent requirement, from 2021, which has seemingly identical convoluted evaluation factors. No way this 190 page+ solicitation was written in 2020. If that office is anything like every other contracting office, the original document was written many years ago and is passed around the office whenever one of these sorts of requirements comes up.

I am sympathetic. The last time the solicitation had (probably) 200+ pages and eight or so factors, was protested, GSA fumbled its corrective action and ultimately ended up paying protestor costs. The final documents that came out of that mess surely include hard-won lessons. Why discard that and start anew? Even if the CO thinks she can probably strip out six of those evaluation factors and slim the solicitation down to 10 pages, her management, which remembers paying the lawyer fees last time, will immediately shoot her down, so why bother? Just change the names & dates, update the clauses, and move on.

Edit: This story is why things don't change and why people don't learn. I am sympathetic to my imagined CO, but not in agreement.

CW Government Travel, Inc.--Costs | U.S. GAO

(deleted)

21 hours ago, Vern Edwards said:

he last thing I would want to do is work in a "contracting squadron" or "contracting agency" or staff. I would want to work for, report to, and be evaluated by the man or woman who ultimately needs the product or service.

Years ago, I was an Army officer who had some COR duties while deployed. As I was responsible for base security, I was the COR for related contracts (3rd country guards, perimeter maintenance, interpreters, fancy security devices, etc.). Our base security contracts were ... satisfactory. The Contracting Office was AF. I visited their base sometimes. You know who had exceptional base security contract performance - fences maintained, barriers and bunkers trash-free, well-disciplined contract guards, the latest fancy devices scanning and detecting threats? The base home to the COs who ran the base security contracts.

2 hours ago, FrankJon said:

I'm curious to know whether anyone has experience using PRISM for workload tracking. My office is in the process of converting from our trusty Excel-based trackers to PRISM, and it's not going well -- big gaps in the data we need, wonky presentation, and extreme latency.

Yes. A topic near to my heart. Your PRISM version may be very different from mine, and PRISM is rapidly changing, so not all of this rant may apply.

PRISM (my version and at least some others I know of) uses requisitions (called requisitions in PRISM, but goes by different names, it is the funding document) to track requirements This is PRISM's fatal flaw. Funding Requirement.

I suppose If your office typically has a single requirement with a single requisition that arrives at the beginning of the process, and "workload" means "Preaward-Only" then PRISM is technically acceptable for workload. If not, PRISM is bad, don't use it.

PRISM - at least my version - has minimal to no workload tracking functionality beyond requisitions. Most likely you will need to continue to track workload outside of PRISM and then reconcile the two sources.

Even this most elemental of tracking mechanisms fails if funding (requisitions) is moderately complicated.

Disclosure: I work within HHS, where PRISM is called HCAS. I've said all of this (and more, this is the short list) on the record to HCAS PMs, who will probably read this and disagree with me.

6 hours ago, Vern Edwards said:

What data elements should you track?

Contract data is useful, but what we are doing is people.

CORs. We lost hundreds of CORs, most of our best CORs, and had perhaps a thousand active contracts change CORs (compared to ~100 for prior years). Many brand-new CORs. CORs abruptly assigned to complex ongoing contracts, or to contracts that had no COR for months, etc. Critical COR shorts in many domains - IT, research, facilities, cost contracts, etc. One COR Level 2 has >90 contracts. We've got a COR Level 1 with 40 assigned to them. Few of them understand the nuances of the FAC cert process.

So, we now are building a lot of COR tracking. Something that is new. Whose got the certificates, much more detail about certification and training status, who can do what, who has capacity, etc. As little of this is or can be automated, it's a far high effort for a pretty basic output compared to anything you can do with financial or acquisition data.

Civilian GVT - Not Classified

BLUF - We use Microsoft 365 products (Excel, Power BI), not because they are the best, but because people know how to use them.

Basic individual data needs or ad-hoc reporting is typically Excel files.

Tracking - and other types of sophisticated information that is used by many people - is centrally administrated and communicated out via dashboards and metrics. We primarily use PowerBI, since it is (relatively) easy to use. Some science/tech groups within agency use Tableau, which they are more familiar with. Agency has some extremely power Oracle reporting tools that I detest and never use.

Our data is simple - hosted in private .gov cloud, isn't classified, doesn't need custom permissions, isn't confidential (like, no source selection info, no detailed invoice/cost info). Despite our relative simplicity, data management still manages to be the hardest part. Centrally controlled data is an effort, but better than the alternative of distributing out the raw data and having everyone use Excel to make up their own undocumented, customized, and conflicting metrics.

On 11/11/2025 at 10:37 AM, C Culham said:

the beginning of the time the contractor is responsible for performing the work.

I am honestly unsure of the correct way of thinking about 'total period of performance' - so I default to the more generous definition. For FAR 6.3 purposes, the time after award, but before the requirement is being met, would not count. I suppose this is similar to the bona fide needs lead time exception. Definitely could be wrong though.

Consider a very common thing in my agency - a gas chromatography system. A commercial scientific instrument which (often) needs several weeks of configuration & calibration, and annually renewed software, to be operational. Does the PoP start when the box is delivered on January 1 - or February 2, when it's switched on and starts working - or January 15, at acceptance when the on-site OEM install tech and agency lab guy sign off on it? Genuinely don't know if there is a definitive answer here.

The answers to your questions depend on the governing FAR Parts. Many of your problems go away if this IT is commercial (Part 12), and it is exempt from FAR Part 6 (Part 13).

The hardware delivers ~120 days after award and the software license/support term begins at hardware acceptance, even a 12-month license term would extend the contract beyond 12 months from award, exceeding the FAR 6.302-2 limit unless we obtain HCA approval.

[Correct me if I am wrong wifcon]:

I think a reasonable person would determine that the lead time between delivery of a product, and the start date when the software needed to use the product is turned on, does not count towards the one year "total period of performance".

Typically, in my agency, a contract for this would have two Line Items (we don't do sub-CLINs). CLIN 1 for product with a delivery date and $. CLIN 2, not separately priced, for the software.

If the overall requirement is for the product, and you have to categorize the entire thing as a single value- like PSC for category management, or NAICS for non-manufacturing rule , or whatever - use the product's value.

Often the buyer cannot 'shorten the PoP.' Many software-sellers cannot or will not do this. Typical response: "Sure, government customer, you can limit your PoP to 274 days as you like, just uninstall our software, but the contract will state, and you will pay for, 365 days, just like all our other customers." A reason to know if this is commercial.

If you can separate out the software, and buy it separately through normal procedures, without FAR 6.3, you would of course want to do that.

So best value is the answer to the question "How do (or should) buyers, with multiple criteria, decide what to buy?" Answer: By selecting the proposal that represents best value.

I propose that 'best value' is simply the federal acquisition application of an enormously rich and influential school of economic thought - decision theory. Many Nobel prizes have been won by economists studying this topic. At its broadest, decision theory is the study of how actors make economic decisions.

The pioneers of decision theory, Von Neumann & Morgenstern (1940's), would define best value as the (source selection) decision that maximizes expected utility under constraints.

So Best = Maximum

Value = Utility

Before you object to me kicking the can down the road when it comes to defining 'value' I will quote the great Jeremy Bentham..

By utility is meant that property in any object, whereby it tends to produce benefit, advantage, pleasure, good, or happiness, (all this in the present case comes to the same thing) or (what comes again to the same thing) to prevent the happening of mischief, pain, evil, or unhappiness to the party whose interest is considered: if that party be the community in general, then the happiness of the community: if a particular individual, then the happiness of that individual.

14 hours ago, formerfed said:

demonstrated abilities rather than written dreams

I agree with that. For a large risky IT project - which consolidating >100 separate HR systems into one, for ~2,000,000 employees, in 18 months, after having fired the staff of the agencies upon which the success of this project depends, is indeed - the thing that matters is what the offeror has done.

A good 1102 CE exercise is to have student/class pick some very familiar software (like MS Word, Gmail.) and attempt to describe it in a "technical approach" style and notice how woefully that compares to just showing how the software works. Would you, as an evaluator, know more about MS Word's ability to do what you want after 10 minutes of using MS Word, or after having read a 100-page pdf about MS Word?

18 hours ago, formerfed said:

It is exceedingly complex and will force individual agency HR processes to adapt and revise many unique agency practices.

Good: 119 HR systems is too many. Consolidation seems like a good idea.

Bad: It is the intent of government for all agencies to go live by July 4, 2027.

Completing the transition for the entire federal government - both civilian and military - in 18 months is impossible. Everyone involved knows this. That this ludicrous goal made it into the solicitation is a very bad sign.

Terrible. FFP. Given the complexity, large dollars and ludicrously short designed-to-fail delivery schedule, there are very high risks of many types of bad to catastrophic failure, which the contractor will be nominally absorbing. So this is going to cost...a lot. A FFP contract is going to be galactically expensive (now or later). The GVT cannot actually transfer risk of a mega-project like this onto the contractor, so it's getting the costs of FFP, but not the benefits.

The main use case is when you have a requirement that can be met by some non-MAS IDIQ, but isn't a good fit for an order. An indefinite delivery vehicle (FPDS's term, not mine) would be better. Under the old FAR, you have three options. First, use an MAS to establish a BPA, even if the other IDIQ is otherwise superior. Second is to wield the innovation hammer and beat your indefinite requirement into a definitive order - usually with lots of dubious options, bloated scope, and forgoing future competition (see below). Third is to do an IDIQ. Now, you have a fourth and far superior option - the IDIQ-BPA.

Consider the Best-in-Class sources. OASIS, EIS and NASA SEWP (to name but a few) are all big multiple award IDIQs that effectively issue second-tier IDVs which have been beaten into something which I suppose meets the legal definition of an 'order.' So yes, this can be done - and has been done for years. But just allowing a second-tier IDV and doing away with the subterfuge and complexity (aka 'innovation') would be much better.

Generally, the self-administrated IDIQ is the last resort, anything else is preferable. There are some agencies that are awarding their own IDIQs which could have been an MAS BPA, but not too many of them. This isn't common, or a big problem. At least according to a cursory look at FPDS data. Also according to common sense - IDIQ's more effort than a BPA, for basically the same result. People aren't doing this for fun, or to spite the FAS.

3 hours ago, FrankJon said:

second-tier IDIQ

Having read the relevant USC and GAO case, I agree @FrankJon that a BPA issued against a multiple-award IDIQ is exactly the 'second-tier IDIQ instrument' which will remain inconsistent with statute regardless of what's stated in FAR 16.5.

From B-411699; B-411796: The FBI’s contemplated award of a 5-year second-tier IDIQ instrument to a single contractor is inconsistent with the requirements of the applicable statutes and FAR provisions regarding what constitutes a “delivery order.” Those requirements are, at a minimum, that the delivery order be defined as to quantity, place of delivery and schedule. In essence, the two orders contemplated under these RFPs will deprive all the other TacCom contractors of a fair opportunity to compete for each of the delivery orders that will be issued in the future ...

How was statement true then, but is not true now, given the applicable statutes haven't changed?

Caveats: 1) I am not an attorney, and definitely not a federal contract attorney, nor an expert practitioner. 2) I think second-tier IDIQ instruments are a great idea, and the prohibition against them is red tape.